Instructions to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Notebooks
Google Colab
Kaggle
Local Apps Settings

How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Use Docker

docker model run hf.co/him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

LM Studio
Jan

vLLM

How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Ollama
How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with Ollama:
```
ollama run hf.co/him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M
```
Unsloth Desktop

How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Configure the model in Pi

# Install Pi:
npm install -g @earendil-works/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Docker Model Runner
How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with Docker Model Runner:
```
docker model run hf.co/him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M
```

Lemonade

How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Run and chat with the model

lemonade run user.Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP-Q4_K_M

List all available models

lemonade list

Hermes Agent

How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Run Hermes

hermes

Atomic Chat

OpenClaw

How to use him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP with OpenClaw:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama serve -hf him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M

Configure OpenClaw

# Install OpenClaw:
npm install -g openclaw@latest
# Register the local server and set it as the default model:
openclaw onboard --non-interactive --mode local \
  --auth-choice custom-api-key \
  --custom-base-url http://127.0.0.1:8080/v1 \
  --custom-model-id "him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP:Q4_K_M" \
  --custom-provider-id llama-cpp \
  --custom-compatibility openai \
  --custom-text-input \
  --accept-risk \
  --skip-health

Run OpenClaw

openclaw agent --local --agent main --message "Hello from Hugging Face"

Qwen3.8-Flash-Next-Uncensored Q4_K_M (Integrated MTP)

An integrated MTP (Multi-Token Prediction) GGUF build of the abliterated Qwen3.8-Flash-Next-Uncensored model, quantized to Q4_K_M with the MTP draft head merged into the same 4-shard split — no sidecar file needed.

Model Details

Property	Value
Base model	`Qwen/Qwen3.8-Flash-Next`
Abliteration	`orcarouter/Qwen3.8-Flash-Next-Uncensored`
Quant	Q4_K_M (main trunk) + Q4_K_M (MTP head)
Split	4 shards (3 trunk + 1 MTP), integrated
Total size	~113.7 GiB
Context	262K native
Architecture	qwen4exp (Gated DeltaNet + QSA + HyperConnections + PLE)
MTP	1 draft layer, integrated as `blk.48`

Usage (llama.cpp with qwen4exp support)

./build-vulkan/bin/llama-server \
  --model Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP-00001-of-00004.gguf \
  --flash-attn on \
  --spec-type draft-mtp \
  --spec-draft-adaptive \
  --spec-draft-n-min 0 \
  --spec-draft-n-max 7 \
  --spec-draft-p-min 0.75

What's inside

This repo contains the MTP head tensors fused into the main GGUF:

blk.48.nextn.* — MTP embedding/hidden projections + norms
blk.48.nextn.hc_head_* — draft head output mixer
Full 48-layer trunk with nextn_predict_layers = 1

⚠️ Disclaimer

This model is an abliterated (refusal-removed) build. It will comply with harmful, unethical, or illegal requests the original Qwen3.8-Flash-Next would refuse. Released strictly for legitimate research — interpretability, AI-safety / refusal-mechanism study, red-teaming, and robustness evaluation. You assume full responsibility for how you use it and everything it generates; add your own safety and moderation layers before any deployment.

License

Qwen Community License 1.0 (see LICENSE), inherited from the base model Qwen/Qwen3.8-Flash-Next. Abliteration and quantization do not change the underlying license obligations.

Note: per the Qwen Community License, if you operate a Model as a Service or AI Work Assistant business commercially, you must obtain a separate license from Qwen before using this model or its derivatives for commercial purposes. See LICENSE for full terms.

Downloads last month: 226

GGUF

Model size

180B params

Architecture

qwen4exp

Hardware compatibility

4-bit

Model tree for him0413/Qwen3.8-Flash-Next-Uncensored-Q4_K_M-MTP

Base model

Qwen/Qwen3.8-Flash-Next

Quantized

(172)

this model